HERA, THE SPHINX?

In a spirit of inquiry I invite your attention anew to one of the most familiar fragments of Greek sculpture, the face which has been called the face of the cult statue of Hera from her temple at Olympia ever since it was so identified by the excavators in 1870 (Plate XII, Figs. 1-3).' 1 propose to discuss this identification once more because I think that the statue of Hera was something quite different, and that there is another explanation of the fragment in question, at least as plausible as the interpretationas Hera. The block of limestone is 0.52 m. tall, and consists of a face with the left ear projecting at right angles, with a broken mass behind and below the ear. It has neatly worked curls across the forehead, a band over the hair, straight hair diverging from the crown of the head, and on the very center of the head, a kalathos. On the right side the face is preserved as far back as on the left but there is no ear on the right side. This curious asymmetry has never been explained, nor has there been a convincing explanation of the lump which appears at the base of the headdress on the same side as the ear, without there being any corresponding lump on the other side. As soon as this face was found it was claimed as the head of the cult statue from the temple of Hera for the following reasons: first, the head is too large to be any- thing but a goddess, and the appropriate goddess is Hera, who had a large, ancient and important temple on the grounds; second, the material, a limestone, is the same as the base in the temple on which a statue or statues stood; third, that the head was found not very far from the temple; fourth, that a limestone figure could not stand out-of-doors; fifth, and this is the important argument, that Pausanias decribed a "rude " statue of the seated Hera within the temple. But let us examine the text of Pausanias (V, 17, 1-4). The section opens with the statement that in the temple of Hera is a statue of Zeus, and that he stands beside the seated goddess, and that the workmanship of these statues is rude. Pausanias con- tinues with descriptions of statues of the Seasons and of Themis; of five Hesperides, made by Theokles; of Athena, made by Medon, and of Kore, Demeter, Apollo and Artemis; then of Leto, Tyche, Dionysos, and a winged Nike. He proceeds (Frazer's translation), " The images I have enumerated are of ivory and gold [ra pev 8r; KarEL- in Xeyl vaE(rtv EXEEcav^roKai Xpva-ov]. But afterwards they dedicated other images the Heraeum: Hermes bearing the babe Dionysus, a work of Praxiteles in stone; and a bronze Aphrodite. ... A gilded child, naked, is seated before the image of Aphro-

1 Ausgrabungen zu Olympia, IV, pp. 13 f., pis. XVI f.; Olyvlpia, Ergebnisse, III, pp. 1-4, pi. 1.

American School of Classical Studies at Athens is collaborating with JSTOR to digitize, preserve, and extend access to Hesperia ® www.jstor.org 354 DOROTHY KENT HILL dite. . . . Hither were brought from the so-called Philippeum other statues of gold and ivory: Eurydice, Philip." (The restora- tions at the end were made by Buttmann.) Now an honest and fair reading of this passage demands that all the statues, including the Hera and unless specified to the contrary, were of gold and ivory. The Hermes, the Aphrodite and the child are care- fully singled out as of other material. There is one clear statement that the statues mentioned in the first part of the chapter are of goldgend and ivory, and toward the there is the return to the subject, in the mention of other statues in these materials. The implication is clearly that the Heraion was a sort of storehouse for statues in the fine expensive media, with only a few ordinary ones added. Apparently there are two reasons why this chapter has not been accepted at its face value by archaeologists. First, Pausanias elsewhere mentions some of the same statues in their first settings, with works which he specifies as of cedar wood. It there- fore has seemed that he meant that the pieces in the Heraion were also cedar. Let us read the passages. In VI, 19, 8, describing one of the treasuries of the Epidamnians he says, " It contains a representation of Atlas upholding the firmament, and another of Hercules and the apple-tree of the Hesperides, with the serpent coiled about the tree. These are also of cedar-wood and are works of Theocles, son of Hegylus. The Hesperides were removed by the Eleans, but were still to be seen in my time in the Heraeum." Of this group, Pausanias describes the male participant and the tree with the serpent as of cedar. But he does not say that the female figures in the group were of cedar. Nor does he give any explanation of the splitting of the group. But since he has previously told us that the Hesperides were of ivory and gold, he does not think it necessary to go into explanations here. Apparently they were moved because early chryselephantine figures were being collected in the Heraion. Farther along in the same chapter (VI, 19, 12), Pausanias says, " The people of Megara, near Attica, built a treasury, and dedicated offerings in it, consisting of small cedar-wood figures inlaid with gold, and representing Hercules' fight with Achelous. Here are represented Zeus, Dejanira, Achelous, and Hercules, and Ares who is helping Achelous. Also there was formerly an image of Athena, because she was an ally of Hercules: but this image now stands beside the Hesperides in the Heraeum. . The treasury in Olympia was made by the Megarians years after the battle, but they must have had the votive offerings from of old, since they were made by the Lacedaemonian Dontas, a pupil of Dipoenus and Scyllis." Pausanias in this passage does not say that Athena was of cedar and gold, but he does imply it. In a few moments we shall see that this implication is not irreconcilable with his previous statement that Athena was of gold and ivory. When:I spoke of an honest and fair reading of Pausanias, I was not flattering a revolutionary reading of my own. Ovierbeckin his An4tikeSchriftquellen zur Ge- schichte der bildenden Kiinste, published in 1868, grouped the three passages from. HERA, THE SPHINX? 355

Pausanias which I have quoted under the heading, "Anfange der Goldelfenbein- bildnerei."2 He divided V, 17, 1-4, putting the section about the Hesperides under the artists Hegylos and Theokles, the rest of the section, everything from r'7 "Hpas 8E e(Trr Ev TO)vao to Era eV KKaTrELXEYLLevaE-'rrTv EXE'aavros Kat Xpv-cov, under Dontas and Dorykleidas (equating Medon and Dontas). If Overbeck read the Greek this way, it is certain that the Greek means that these statues were chryselephantine. Two years after the publication of his book, the stone head was found at Olympia and threw dust in the eyes of scholars. Is it unfair to suggest that the wish to have dis- covered the cult statue was father to the thought ? The second reason why Pausanias' statement in Book V, chapter 17, was not taken seriously from the time of the discoveries at Olympia until the present is that until recently it seemed absolutely impossible and absurd to all archaeologists. Who ever heard of any such number of chryselephantine statues, and such early ones, all in one place? Recently there was discovered at Delphi a cache of fragments of chryselephantine statues, three as large as life and one as early as the beginning of the sixth century B.C., showing that such works did exist and were kept together.3 The cache at Delphi was made by shovelling the contents of just such a museum as the Heraion into a trench in the ground, followTinga fire at the end of the fifth century B.C. These Delphi statutes explain Pausanias' apparent confusion. There is not much difference between these early chryselephantine statues made of wood with gold and ivory inlays and attachments and the cedar statues with gold inlays which he described. All were constructed in the same way. Some had much ivory, others less, still others none. The prettiest, the most spectacular, statues were moved to the Heraion although they belonged to groups, to stand beside the cult statue which had always been there. These had lots of gold and ivory, and Pausanias refers to them as made of these materials. The others which exposed more wood and had little or none of the precious materials were left in their original positions. He calls them cedar-wood, or cedar- wood inlaid with gold. Perhaps even if the pieces had not been found at Delphi one could have had an inkling of Pausanias' meaning from a sentence that comes in VI, 19, 10, between the references to the groups which were moved to the Heraion. He speaks of a statue of Dionysos with face, hands, and feet of ivory. He does not tell us what the other parts were. But does anyone doubt that they were of wood? Here we have another indica- tion that such statues existed, and an example of Pausanias' habit of mentioning only what he thinks interesting about a statue. He did not write for us, but for people who could see the originals. Many modern guide books are unintelligible if read away

2 pp. 56-57. 3 A.J.A., XLIII, 1939, p. 696; Gnomon, XV, 1939, pp. 527 f.; B.C.H., LXIII, 1939, pp. 86 ff.; Rev. Arch., ser. 6, XIII, 1939, pp. 263 f.; London Illustrated News, July 29, 1939, pp. 202 f. Only in Rev. Arch., 1939, pp. 263 f. is it stated that these statues had iron armatures. 356 DOROTHY KENT HILL from the scene because they assume a knowledge which only the visitor at the site can have. Some of the Delphi statues are large, but none colossal. We should expect a very large cult statue in the Heraion, since it is a very big temple. But is not the technique of the Delphi figures just the thing for a large statue? A figure built up of parts attached to wood, to a wood core which itself can be of many parts: is this not easier for a primitive than carving a monolith? Leaving all debatable matters aside, the one thing which Pausanias does not say is that the Hera was of stone. If we accept the great stone face as Hera, we must not think that we have his authority for this identification. Actually, once it is admitted that Pausanias does not mention a stone Hera, all reasons for assigning the stone face to the cult statue crumble into dust. That the material is the same as the statue-base means only that it is a local stone, available for all purposes. Then, that the head was found fairly near the temple means little, for it was not found in the temple. That a limestone statue had to stand indoors is not convincing proof that this one stood in the Heraion. It is not even true, for limestone was used out-of-doors. The piece in question may have stood under other protection or it may have been made to stand out-of-doors, in which case its relatively good surface would mean merely that it did not stand for a long period. As against the remaining reason, that the head is too large to be anything but a cult statue, I presume to offer another restoration of the head. I do not consider it final, but to me it seems more probable than the interpretation as a goddess, and it explains the uneven ears. The ear which is preserved is in the wrong position on the head. It is too far forward. The correct position is farther back, in the part now destroyed. We may suppose that the right side showed the ear correctly, back on the head, where the blow which severed the head cut it off. Now placing the ears too far forward is not characteristic of Daedalic statues generally, but of those which have locks of hair falling in front of the shoulders. The straight locks of hair falling before the shoulders in a near vertical plane make it obligatory to show the ears (if one does show them, not conceal them with hair) farther forward than nature asks. Since this hair arrangement is quite common in the seventh century, most of the statues of the period have ears in a forward position. However, there are pieces of that period which show that the correct position of the ear was known, and that it was placed there when the headdress did not require the forward position.4 Also, a bronze statuette of a century later resorts to the same position because the hair is arranged in this way.5

4 Jenkins,Dedalica, pl. VI, la and 3. 5 Artemis in Berlin. Gerke, Griechische Plastik in archaischer und klassischer Zeit, pi. 88. HERA, THE SPHINX? 357

Further, we may notice that it is quite usual for statues of the type which we have mentioned to turn the ears out at right angles to the head. The best example is a large stone statue from Eleutherna in Crete.6 Because there is a plane behind the ears, it is natural to flatten the ears against the plane. To show the ear to the spectator who stands before the statue, this suits the taste of the early Greek artist, a taste which he shares with his Egyptian predecessor, but which he indulges, as far as we can tell, only when he has a flat surface against which to plaster the ear. An ear standing out at right angles to the head without any hair mass behind it, that is something hard to imagine. The left ear of the "fHera," then, we may be sure was combined with locks of hair which fell before the shoulder. One can almost follow the outline of the damaged hair mass around the ear. A comparison of the left side of the head of the statue from Eleutherna with the left side of " Hera " (Plate XII, Fig. 2) should, I think, convince anyone that the two were alike. The right ear was not in the corresponding position on the right side. This sug- gests that on the right side the hair mass was not pulled before the shoulder, pulling the ear forward and out. The reason for the disparity must be sought beyond the head in some asymmetry of the body. What human-headed figure did not show both sides alike? Why, the sphinx who, quietly seated on her haunches, turns her head over her shoulder so that you may see her face in front view at the same moment that you see her lioness' body in all its beautiful profile. The maker of a sphinx in this position must effect a transition between head and body, and his transition is likely to have some influence upon hair and ears. In at least four known cases there was adopted a solution of letting the long tresses which fall from one side of the head, the side turned toward the wing, be pulled forward in front of the wing, while those on the other side fall before the part which is the lioness' chest or the woman's other shoulder. These are: a small limestone sphinx from Sparta,7 a stone sphinx from Marion in Cyprus, now in the Louvre (Plate XII, Fig. 4), and its companion piece, headless, in Berlin,8 and a marble Attic grave sphinx in the Metropolitan Museum of Art (Plate XIII, Fig. 5).9

6 Jahreshefte, XII, 1909, p. 245, figs. 119, 120; Jenkins, op. cit., pl. VIII, nos. 1, la; IX, 1; X, 2. Other good examples are the small terracotta figurines, ibid., pl. I, 1 and 2; Grace, Archaic Sculpture in Boeotia, figs. 29-30; also the large terracotta sphinx heads from Thebes and Kalydon, Payne, Necrocorinthia, pl. 49; and the famous stone Nikandre. 7 B.S.A., XIV, 1907-8, p. 25, fig. 10. Only one lock is visible in the photograph; it falls over the wing. 8 B.C.H., XVIII, 1894, pl. VII; Perrot and Chipiez, Histoire' de I'art dans l'antiquite, VIII, p. 329, fig. 142; Reinach, Rep. de la Stat., II, 705, 5; Ohnefalsch-Richter, Kypros, pl. XXVII, figs. 1, 2. Metropolitan Museum of Art, no. 11.185. A.J.A., XLV, 1941, pp. 159 -f.; Bulletin Met. Mus. Art, XXV, 1940, pp. 178-180. E. Douglas Van Buren deduces a terracotta sphinx from Olympia 358 DOROTHY KENT HILL

Let us examine the New York sphinx. This entrancing creature shows us every- thing from the front. Her lioness' body with tail curled toward us, her face and her full curled wing are all visible from the front. The sculptor has had difficulties with the hair. His solution has been to draw the locks before the wing on one side and to let them fall before the chest from the other side of the head. Therefore, on the side toward the wing there is a protruding mass of hair against which the ear is projected; while on the other side he could and did let the ear lie flat against the head. Difference of position of the ears he has avoided by curving the locks which are pulled forward into an arc, thus giving space for the ear, and by turning the head just slightly away from the profile position of the body. Another solution of the problem is exemplified in Plate XIII, Figs. 6-8.1? The sculptor of this sphinx let the hair on the side toward the wing fall behind the back wing and that on the other side of the head fall before the chest; therefore, although neither ear projects, the inner one, the one toward the wing, is placed farther back on the head. There are, of course, still other solutions. One is to cut the hair off at shoulder length, as in Fig. 9, Plate XIV.11 But enough has been said to show that uneven ears are not unlikely on sphinxes.12 There is no parallel for the " IHera " face among the extant sphinxes. But if we imagine a very primitive artist setting out to make a sphinx with head turned over the left shoulder (the position of the Marion sphinx) we see that he might make just such a face as was found at Olympia. On the side of the head toward the wing, he would pull the hair forward to lie on the front of the wing; therefore, the ear had to be placed well forward and turn out. On the other side the locks of hair would fall straight down before the chest and not protrude; therefore, the ear could lie flat against the head and be where one would like to have it. If the head were split in two from side to side, one ear might be cut off while the other remained. Let us now look at the Olympia piece to see whether it might be the face of a sphinx. As far as the physiognomy is concerned, we will get our answer from a com- parison made as long ago as 1909 by Loewy.13 In discussing the types of early sculp- ture, he had occasion to group on two adjoining pages photographs of the head in with the same hair arrangement, Greek Fictile Revetments in the Archaic Period, p. 174, no. 18. But the hair and necklaceupon whichthis assumptionrests are part of a figureof Nike; cf. Olympia, Ergebnisse,III, p. 40. 10Walters Art Gallery, no. 54.770. Height, 0.072 m. Sale Catalogue,Lambros and Dattari Collections, 1912, no. 214, pl. XVIII; Reinach, op. cit., V, 408, 2; Payne and others, Perachora, p. 135, fig. 19, no. 2. 1Walters Art Gallery,no. 54.1078. Height, 0.084 m. Sale Catalogue,Lambros and Dattari Collections,no. 215, pl. XVIII; Reinach,op. cit., IV, 405, 1. 12 A head in the ClevelandMuseum which has always been called a sphinx has one ear higher than the other, Bulletin of the ClevelandMuseum, XV, 1928, pp. 189, 191 ff. 13 Jahreshefte,XII, 1909, pp. 268 f. HERA, THE SPHINX? 359

question, the sphinx of Spata, the Nike of Delos, a head from the Athenian acropolis, the Kore of Lyons, an early male face, and the famous calf-bearer. A glance at these illustrations should convince anyone that it is absolutely impossible to distinguish the features and expression of an early sphinx from those of a human. Now, to the headdress. It is the kalathos, the basket surrounded by leaves. Figure 9, Plate XIV, illustrates a bronze sphinx wearing the kalathos with leaves.14 So we see that there is no feature or attribute of the Olympia head which is incompatible with its inter- pretation as a sphinx. The reconstruction should take into account the stub at the base of the headdress on the left side. Something touched the head here, and it was something small, for the details of the head are worked to its very edge at top, bottom and front. Treu explained it as part of the veil of Hera.15 I think that the tip of the wing or a strut connected with the wing touched the head at this point. Having the wing connected with the head would greatly strengthen the construction, making breakage of the wing much less likely. There is no parallel for this construction in stone, as far as I know, but there is a class of flat terracotta sphinxes of late Corinthian manufacture which are made in just this way.16 Some of them have a short strut from the head to the wing (Plate XIV, Fig. 10) some made in other but very similar moulds let the wing with its curved top actually rest against the head just below the headdress. If this con- struction was found advantageous for small terracottas, how much more so would it have been for a giant stone sphinx! Returning to the argument of size, one must admit that the size requires the face to come from a cult statue if it comes from a human statue at all. But once the possi- bility is recognized that it may be from an animal figure, the size is no longer a factor in the argument. It may not be altogether beside the point to remark that the great size of the face is not as remarkable in a sphinx as in a human figure; for the total size of the sphinx would not be as great as the total human statue. A seated sphinx with this face might be less than two meters tall, but a seated human figure would have to be two and a half or more. I have no evidence for assigning an exact location in the Altis to the sphinx that I have hypothesized, and will not attempt to say what kind of monument it adorned. Its equal in size was the marble sphinx dedicated by the Naxians at Delphi.17 It will be salutary to remember,before we reject the interpretation of the Olympia fragment

14 A terracotta head found at Olympia has the kalathos, rendered exactly as on the great stone face, Olympia, Ergebnisse, III, p. 35, figs. 35 and 36, pl. VII, 1. It was suggested that this was a Hera because of the resemblance to the other. But what could be more usual than a sphinx head in terracotta ? 15 Olympia, Ergebnisse, III, p. 1. 16 Payne and others, Perachora, p. 234, no. 194, pl. 101; Jacobsthal, Die melischen Reliefs, p. 89, pl. 66 b; Winter, Die Typen der figurlichen Terrakotten, I, p. 229, no. 9 (many examples). 17 B.C.H., XVII, 1893, p. 615; Fouilles de Delphes, IV, pt. I, p. 42. 360 DOROTHY KENT HILL as a sphinx' head, that the head of the Delphi sphinx was called an Apollo when it was first discovered, and that a terracotta sphinx head from Kalydon was called an Artemis.18 I do not claim to have proved that "Hera" is really a sphinx, but to have sug- gested it. This is the first explanation which has ever been given for the unevenness of the ears. It seems to me that with the accumulatedknowledge of recent years at his finger tips, a modern excavator would immediately assume that the head came from a sphinx, not from a goddess. Would it not be worth while to try out this assumption and see where it leads us in knowledge and understanding of archaic art? 19 But I hope that I have shown that the other assumption, that this is the cult statue from the temple of Hera, is wTithoutfoundation. In the light of the Delphi finds we ought to accept Pausanias' statement of what he saw in this temple: twenty-two or more chryselephantine statues large and small, including the cult statue, most of them very ancient, and along with them one marble statue, one bronze, and one gilded bronze. Is this not a glorious picture? On the terrible day of destruction they all were looted for their valuable material, not a bit of ivory, gold, bronze, or gilded bronze being spared, but the marble statue of Hermes was left for the excavators because it had no intrinsic value. DOROTHY KENT HILL THE WALTERS ART GALLERY

18 Poulsen and Rhomaios, "Erster vorlaufiger Bericht iiber die danisch-griechischen Aus- grabungen von Kalydon" (Det Kgl. Danske Videnskabernes Selskab, Hist. fil. Med., XIV, 3), 1927, p. 29. 19Various attempts were made to connect sculptural fragments from Olympia with the " Hera." They were unsuccessful. My efforts to connect them with the sphinx have failed as well. They are: some huge fragments, once called parts of the throne of Hera, though this assignation was repudiated in the final Olympia publication; a bit of the mane of a huge limestone lion; and a bit of human hair. Only on the occasion of a visit to Olympia can anything be decided about these. PLATE XII

Fig. 1. Head of Hera at Olympia Fig. 2. Profile of Head in Figure 1 From a Cast

Fig. 3. Profile of Head Fig. 4. Sphinx from Marion, Cyprus in Figure 1 After Bulletin de CorrespondanceHellenique

HILL: HERA, THE SPHINX? PLATE XIII

Fig. 5. Marble Sphinx from Grave Fig. 6. Bronze Sphinx. Walters Art Monument. Metropolitan Gallery Museum of Art

Fig. 7. Diagonal View of the Sphinx in Fig. 8. Diagonal View of the Sphinx in Figure 6 Figure 6

HILL: HERA, THE SPHINX?